Estrogen represses Tgfbr1 and Bmpr1a expression via estrogen receptor beta in MC3T3-E1 cells

MC3T3-E1 is a clonal pre-osteoblastic cell line derived from newborn mouse calvaria, which is commonly used in osteoblast studies. To investigate the effects of estrogen on osteoblasts, we treated MC3T3-E1 cells with various concentrations of estrogen and assessed their proliferation. Next, we performed RNA deep sequencing to investigate the effects on estrogen target genes. Bmpr1a and Tgfbr1, important participants in the TGF-beta signaling pathway, were down-regulated in our deep sequencing results. Bioinformatics analysis revealed that estrogen receptor response elements (EREs) were present in the Bmpr1a and Tgfbr1 promoters. Culturing the cells with the estrogen receptor (ER) alpha or beta antagonists 1,3-bis(4-hydroxyphenyl)-4-methyl-5-[4-(2-piperidinylethoxy)phenol]-1H-pyrazole dihydrochloride (MPP) or 4-[2-phenyl-5,7-bis(trifluoromethyl) pyrazolo[1,5-alpha]pyrimidin-3-yl] phenol (PTHPP), respectively, demonstrated that ER beta is involved in the estrogen-mediated repression of Tgfbr1 and Bmpr1a.The chromatin immunoprecipitation (ChIP) results were consistent with the conclusion that E2 increased the binding of ER beta at the EREs located in the Tgfbr1 and Bmpr1a promoters. Our research provides new insight into the role of estrogen in bone metabolisms.


88
Cell culture 89 The clonal murine pre-osteoblastic cell line MC3T3-E1 was used in this study. The cells were 90 thawed from frozen stocks and cultured in 100-mm culture dishes in α-MEM supplemented 91 with 10% (v/v) fetal bovine serum (Gibco), 1% (w/v) antibiotics/antimycotics (the stock 92 solution contained 100 U/mL penicillin G sodium, 100 µg/mL streptomycin sulfate, and 0.25 93 µg/mL amphotericin B, in saline), 1.0 mM sodium pyruvate, 0.1 mM nonessential amino 94 acids, and 1.5 g/L sodium bicarbonate (Gibco) at 37°C in a humidified atmosphere containing 95 5% (v/v) CO2. After establishing cultures from frozen cells, we sub-cultured the cells several 96 times. Confluent cells were detached using 0.25% trypsin in Mg 2+ -and Ca 2+ -free 97 phosphate-buffered saline (PBS) before use. For the treatment of MC3T3-E1 with E2, the 98 cells were seeded in each well of a 24-or 6-well plate with 0.5 or 2mL of culture medium.

99
The cells were incubated in the presence of E2 at different concentrations (0.01, 0.1, 1, and 100 10nM), and each concentration of E2 was maintained for three days. A group that received 101 DMSO served as the control.    119 We used 20μg of total RNA from the control group and the10 nM E2-treated group for the 120 RNA sequencing (RNA-seq) library preparation. The collected mRNA was purified and 121 concentrated using oligo (dT)-conjugated magnetic beads (Invitrogen) before library 122 preparation. The purified mRNA was randomly cut into fragments with the fragmentation 123 buffer. We used the mRNA as a template, with 6-base random primers (random hexamers), to 124 synthesize the first strand of cDNA. We then added dNTPs, RNase H, and DNA polymerase I 125 to the buffer, to synthesize the second strand of cDNA. Lastly, we used AMPure® XP beads 126 (Beckman) to purify the synthesized cDNA. The purification of double-stranded cDNA was 127 performed with end repair and A-tailing. The AMPure® XP beads were used to select the size 128 of the fragments. Finally, the cDNA library was obtained by PCR. Using "sequencing by  high-throughput sequencing platform and obtained high-quality reads. The reads and bases sequenced by the platform were usually considered raw data, and most of them got a Q30 132 score for base quality. DEGs is known as differential expression analysis.  Total cell mRNA was extracted using TRIzol®, according to manufacturer instruction 160 (Invitrogen). We subjected 10μg of total RNA to DNaseI treatment with 1 U DNaseI (NEB).

161
The reaction was carried out at 37°C for 10 min followed by heat inactivation at 65°C for10  Table 1.   Table 1. control groups (Fig. 1). However, in the high concentration E2 groups (1 and 10nM E2),

200
MC3T3-E1 cell proliferation was significantly higher than in the DMSO control groups (Fig.   201 1). These data revealed that high concentrations of E2 promote MC3T3-E1 cell proliferation.

202
In order to obtain robust data and explore the mechanisms of action of estrogen on  Table 2. We found that genes that affect biological regulation, metabolic processes,  (Table 2).  (Table 3). We also noted alterations in the expression of Lrp5, Vangl2, and Rock1, 230 which are involved in Wnt signaling (Table 3). In the cell cycle signaling pathway, we noted  236 We chose the candidate target genes from the bone metabolic signaling pathway. The  (Fig. 3B), Tgfbr1 (Fig. 3C), Rock1 (Fig. 3D), Lrp6, (Fig. 3F), 242 and Sp1 (Fig. 3H) were down-regulated by E2 treatment, whereas Fos (Fig. 3E) and Adra1b 243 (Fig. 3G) were up-regulated. These results confirmed the differential expression after E2 244 treatment of the genes identified by RNA-seq. Tgfbr1gene promoters, respectively (Fig. 4A). To determine if E2 repressed the candidate 254 genes via a particular ER, we treated MC3T3-E1cells with 10nM E2, then added the ER 255 antagonist MPP or the ERβ antagonist PTHPP. We then performed qRT-PCR to investigate 256 candidate gene expression. We found that E2 treatment significantly down-regulated Bmpr1a 257 and Tgfbr1, compared to their expression in the DMSO control groups, in the MC3T3-E1 258 cells (Fig.4B-D). MPP did not rescue the E2-mediated down-regulation of Bmpr1a and Tgfbr1 259 (Fig. 4B-D). However, PTHPP treatment rescued the down-regulation of Bmpr1a and Tgfbr1 260 (Fig. 4B-D). Rescue of the candidate target gene expression by PTHPP, but not MPP, revealed 261 that estrogen negatively regulates gene expression via ERβ. 264 We next addressed the potential functionality of the EREs located in the Bmpr1a and Tgfbr1 265 promoters. We performed ChIP experiments using an ERβ antibody to determine if ERβ 266 bound to the EREs in the Bmpr1a and Tgfbr1 promoters. We found that ERβ bound to ERE1 267 and ERE3, but not ERE2, in the Bmpr1a promoter (Fig. 5A). Both of the EREs located in the 268 Tgfbr1promoter were bound by ERβ (Fig. 5B). To investigate if E2 treatment affected the 269 binding of ERβ in the Bmpr1a and Tgfbr1 promoters, we treated the MC3T3-E1 cells with E2 270 for 72h, then performed ChIP, followed by qRT-PCR. E2 treatment increased ERβ binding to 271 ERE1 and ERE3 in the Bmpr1a promoter (Fig. 6C), and ERE1 and ERE2 in the Tgfbr1 272 promoter (Fig. 5D). These data demonstrated that ERβ can bind to the Bmpr1a and Tgfbr1     360 The authors declare that they have no conflict of interest.      Student's t test was performed. *, P < 0.05; **, P < 0.01; NS, not significant.